Index maintenance in distributed availability group

I'd like your opinion about Index Maintenance in Distributed availability groups. In this way, the size of the databases is at the limit of terabytes, and this process is supposed to be transferred to one or more other sites. In your opinion, what best practices should we consider here? These sites are connected synchronously; we should not lose data. Now we want to execute the Rebuild index operation. Considering the type of waits and network bandwidth, etc., what is the best way to manage indexes and perform these processes?